Read font size for runs

1 files changed, 88 insertions, 80 deletions

diff --git a/mammoth/docx/body_xml.py b/mammoth/docx/body_xml.py
index d6ca993..7bb8628 100644
--- a/mammoth/docx/body_xml.py
+++ b/mammoth/docx/body_xml.py
@@ -22,10 +22,10 @@ def reader(
     docx_file=None,
     files=None
 ):
-    
+
     if styles is None:
         styles = Styles.EMPTY
-    
+
     read_all = _create_reader(
         numbering=numbering,
         content_types=content_types,
@@ -41,7 +41,7 @@ def reader(
 class _BodyReader(object):
     def __init__(self, read_all):
         self._read_all = read_all
-    
+
     def read_all(self, elements):
         result = self._read_all(elements)
         return results.Result(result.elements, result.messages)
@@ -79,20 +79,27 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             .find_child_or_null("w:vertAlign") \
             .attributes.get("w:val")
         font = properties.find_child_or_null("w:rFonts").attributes.get("w:ascii")
-        
+
+        font_size_string = properties.find_child_or_null("w:sz").attributes.get("w:val")
+        if font_size_string:
+            # w:sz gives the font size in half points, so halve the value to get the size in points
+            font_size = int(font_size_string) / 2
+        else:
+            font_size = None
+
         is_bold = read_boolean_element(properties.find_child("w:b"))
         is_italic = read_boolean_element(properties.find_child("w:i"))
         is_underline = read_boolean_element(properties.find_child("w:u"))
         is_strikethrough = read_boolean_element(properties.find_child("w:strike"))
         is_small_caps = read_boolean_element(properties.find_child("w:smallCaps"))
-        
+
         def add_complex_field_hyperlink(children):
             hyperlink_href = current_hyperlink_href()
             if hyperlink_href is None:
                 return children
             else:
                 return [documents.hyperlink(href=hyperlink_href, children=children)]
-        
+
         return _ReadResult.map_results(
             _read_run_style(properties),
             _read_xml_elements(element.children).map(add_complex_field_hyperlink),
@@ -107,8 +114,9 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                 is_small_caps=is_small_caps,
                 vertical_alignment=vertical_alignment,
                 font=font,
+                font_size=font_size,
             ))
-    
+
     def _read_run_style(properties):
         return _read_style(properties, "w:rStyle", "Run", styles.find_character_style_by_id)
 
@@ -121,7 +129,7 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
         alignment = properties.find_child_or_null("w:jc").attributes.get("w:val")
         numbering = _read_numbering_properties(properties.find_child_or_null("w:numPr"))
         indent = _read_paragraph_indent(properties.find_child_or_null("w:ind"))
-        
+
         return _ReadResult.map_results(
             _read_paragraph_style(properties),
             _read_xml_elements(element.children),
@@ -133,20 +141,20 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                 alignment=alignment,
                 indent=indent,
             )).append_extra()
-    
+
     def _read_paragraph_style(properties):
         return _read_style(properties, "w:pStyle", "Paragraph", styles.find_paragraph_style_by_id)
-    
+
     current_instr_text = []
     complex_field_stack = []
-    
+
     def current_hyperlink_href():
         for complex_field in reversed(complex_field_stack):
             if isinstance(complex_field, complex_fields.Hyperlink):
                 return complex_field.href
 
         return None
-    
+
     def read_fld_char(element):
         fld_char_type = element.attributes.get("w:fldCharType")
         if fld_char_type == "begin":
@@ -164,24 +172,24 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             complex_field_stack.pop()
             complex_field_stack.append(complex_field)
         return _empty_result
-    
+
     def parse_hyperlink_field_code(instr_text):
         result = re.match(r'\s*HYPERLINK "(.*)"', instr_text)
         if result is None:
             return None
         else:
             return result.group(1)
-    
+
     def read_instr_text(element):
         current_instr_text.append(_inner_text(element))
         return _empty_result
-    
+
     def _read_style(properties, style_tag_name, style_type, find_style_by_id):
         messages = []
         style_id = properties \
             .find_child_or_null(style_tag_name) \
             .attributes.get("w:val")
-        
+
         if style_id is None:
             style_name = None
         else:
@@ -191,9 +199,9 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                 messages.append(_undefined_style_warning(style_type, style_id))
             else:
                 style_name = style.name
-        
+
         return _ReadResult([style_id, style_name], [], messages)
-    
+
     def _undefined_style_warning(style_type, style_id):
         return results.warning("{0} style with ID {1} was referenced but not defined in the document".format(style_type, style_id))
 
@@ -216,31 +224,31 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
 
     def tab(element):
         return _success(documents.tab())
-    
-    
+
+
     def no_break_hyphen(element):
         return _success(documents.text(unichr(0x2011)))
-    
-    
+
+
     def table(element):
         properties = element.find_child_or_null("w:tblPr")
         return _ReadResult.map_results(
             read_table_style(properties),
             _read_xml_elements(element.children)
                 .flat_map(calculate_row_spans),
-                
+
             lambda style, children: documents.table(
                 children=children,
                 style_id=style[0],
                 style_name=style[1],
             ),
         )
-            
-    
+
+
     def read_table_style(properties):
         return _read_style(properties, "w:tblStyle", "Table", styles.find_table_style_by_id)
-    
-    
+
+
     def table_row(element):
         properties = element.find_child_or_null("w:trPr")
         is_header = bool(properties.find_child("w:tblHeader"))
@@ -249,19 +257,19 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                 children=children,
                 is_header=is_header,
             ))
-    
-    
+
+
     def table_cell(element):
         properties = element.find_child_or_null("w:tcPr")
         gridspan = properties \
             .find_child_or_null("w:gridSpan") \
             .attributes.get("w:val")
-        
+
         if gridspan is None:
             colspan = 1
         else:
             colspan = int(gridspan)
-        
+
         return _read_xml_elements(element.children) \
             .map(lambda children: _add_attrs(
                 documents.table_cell(
@@ -270,7 +278,7 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                 ),
                 _vmerge=read_vmerge(properties),
             ))
-    
+
     def read_vmerge(properties):
         vmerge_element = properties.find_child("w:vMerge")
         if vmerge_element is None:
@@ -278,8 +286,8 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
         else:
             val = vmerge_element.attributes.get("w:val")
             return val == "continue" or not val
-    
-    
+
+
     def calculate_row_spans(rows):
         unexpected_non_rows = any(
             not isinstance(row, documents.TableRow)
@@ -289,7 +297,7 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             return _elements_result_with_messages(rows, [results.warning(
                 "unexpected non-row element in table, cell merging may be incorrect"
             )])
-            
+
         unexpected_non_cells = any(
             not isinstance(cell, documents.TableCell)
             for row in rows
@@ -299,7 +307,7 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             return _elements_result_with_messages(rows, [results.warning(
                 "unexpected non-cell element in table row, cell merging may be incorrect"
             )])
-        
+
         columns = {}
         for row in rows:
             cell_index = 0
@@ -310,41 +318,41 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                     columns[cell_index] = cell
                     cell._vmerge = False
                 cell_index += cell.colspan
-        
+
         for row in rows:
             row.children = lists.filter(lambda cell: not cell._vmerge, row.children)
             for cell in row.children:
                 del cell._vmerge
-        
+
         return _success(rows)
-    
-    
+
+
     def read_child_elements(element):
         return _read_xml_elements(element.children)
-    
-    
+
+
     def pict(element):
         return read_child_elements(element).to_extra()
-    
-    
+
+
     def hyperlink(element):
         relationship_id = element.attributes.get("r:id")
         anchor = element.attributes.get("w:anchor")
         target_frame = element.attributes.get("w:tgtFrame") or None
         children_result = _read_xml_elements(element.children)
-        
+
         def create(**kwargs):
             return children_result.map(lambda children: documents.hyperlink(
                 children=children,
                 target_frame=target_frame,
                 **kwargs
             ))
-        
+
         if relationship_id is not None:
             href = relationships.find_target_by_relationship_id(relationship_id)
             if anchor is not None:
                 href = replace_fragment(href, anchor)
-            
+
             return create(href=href)
         elif anchor is not None:
             return create(anchor=anchor)
@@ -358,8 +366,8 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             return _empty_result
         else:
             return _success(documents.bookmark(name))
-    
-    
+
+
     def break_(element):
         break_type = element.attributes.get("w:type")
 
@@ -373,7 +381,7 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             warning = results.warning("Unsupported break type: {0}".format(break_type))
             return _empty_result_with_message(warning)
 
-    
+
     def inline(element):
         properties = element.find_child_or_null("wp:docPr").attributes
         if properties.get("descr", "").strip():
@@ -386,25 +394,25 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             .find_children("pic:blipFill") \
             .find_children("a:blip")
         return _read_blips(blips, alt_text)
-    
+
     def _read_blips(blips, alt_text):
         return _ReadResult.concat(lists.map(lambda blip: _read_blip(blip, alt_text), blips))
-    
+
     def _read_blip(element, alt_text):
         return _read_image(lambda: _find_blip_image(element), alt_text)
-    
+
     def _read_image(find_image, alt_text):
         image_path, open_image = find_image()
         content_type = content_types.find_content_type(image_path)
         image = documents.image(alt_text=alt_text, content_type=content_type, open=open_image)
-        
+
         if content_type in ["image/png", "image/gif", "image/jpeg", "image/svg+xml", "image/tiff"]:
             messages = []
         else:
             messages = [results.warning("Image of type {0} is unlikely to display in web browsers".format(content_type))]
-            
+
         return _element_result_with_messages(image, messages)
-    
+
     def _find_blip_image(element):
         embed_relationship_id = element.attributes.get("r:embed")
         link_relationship_id = element.attributes.get("r:link")
@@ -412,29 +420,29 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
             return _find_embedded_image(embed_relationship_id)
         elif link_relationship_id is not None:
             return _find_linked_image(link_relationship_id)
-    
+
     def _find_embedded_image(relationship_id):
         target = relationships.find_target_by_relationship_id(relationship_id)
         image_path = uri_to_zip_entry_name("word", target)
-        
+
         def open_image():
             image_file = docx_file.open(image_path)
             if hasattr(image_file, "__exit__"):
                 return image_file
             else:
                 return contextlib.closing(image_file)
-        
+
         return image_path, open_image
-    
-    
+
+
     def _find_linked_image(relationship_id):
         image_path = relationships.find_target_by_relationship_id(relationship_id)
-        
+
         def open_image():
             return files.open(image_path)
-        
+
         return image_path, open_image
-    
+
     def read_imagedata(element):
         relationship_id = element.attributes.get("r:id")
         if relationship_id is None:
@@ -443,19 +451,19 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
         else:
             title = element.attributes.get("o:title")
             return _read_image(lambda: _find_embedded_image(relationship_id), title)
-    
+
     def note_reference_reader(note_type):
         def note_reference(element):
             return _success(documents.note_reference(note_type, element.attributes["w:id"]))
-        
+
         return note_reference
-    
+
     def read_comment_reference(element):
         return _success(documents.comment_reference(element.attributes["w:id"]))
 
     def alternate_content(element):
         return read_child_elements(element.find_child("mc:Fallback"))
-    
+
     def read_sdt(element):
         return read_child_elements(element.find_child_or_null("w:sdtContent"))
 
@@ -493,7 +501,7 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
         "mc:AlternateContent": alternate_content,
         "w:sdt": read_sdt
     }
-    
+
     def read(element):
         handler = handlers.get(element.name)
         if handler is None:
@@ -504,12 +512,12 @@ def _create_reader(numbering, content_types, relationships, styles, docx_file, f
                 return _empty_result
         else:
             return handler(element)
-        
+
 
     def _read_xml_elements(nodes):
         elements = filter(lambda node: isinstance(node, XmlElement), nodes)
         return _ReadResult.concat(lists.map(read, elements))
-    
+
     return _read_xml_elements
 
 
@@ -528,20 +536,20 @@ class _ReadResult(object):
             lists.flat_map(lambda result: result.elements, results),
             lists.flat_map(lambda result: result.extra, results),
             lists.flat_map(lambda result: result.messages, results))
-    
-    
+
+
     @staticmethod
     def map_results(first, second, func):
         return _ReadResult(
             [func(first.elements, second.elements)],
             first.extra + second.extra,
             first.messages + second.messages)
-    
+
     def __init__(self, elements, extra, messages):
         self.elements = elements
         self.extra = extra
         self.messages = messages
-    
+
     def map(self, func):
         elements = func(self.elements)
         if not isinstance(elements, list):
@@ -550,18 +558,18 @@ class _ReadResult(object):
             elements,
             self.extra,
             self.messages)
-    
+
     def flat_map(self, func):
         result = func(self.elements)
         return _ReadResult(
             result.elements,
             self.extra + result.extra,
             self.messages + result.messages)
-        
-    
+
+
     def to_extra(self):
         return _ReadResult([], _concat(self.extra, self.elements), self.messages)
-    
+
     def append_extra(self):
         return _ReadResult(_concat(self.elements, self.extra), [], self.messages)
 
@@ -592,5 +600,5 @@ def _concat(*values):
 def _add_attrs(obj, **kwargs):
     for key, value in kwargs.items():
         setattr(obj, key, value)
-    
+
     return obj